Re: [PATCH] LockAcquireExtended improvement

Robert Haas <robertmhaas@gmail.com>

From: Robert Haas <robertmhaas@gmail.com>
To: Jingxian Li <aqktjcm@qq.com>
Cc: andres <andres@anarazel.de>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2024-03-14T13:15:56Z
Lists: pgsql-hackers
On Tue, Mar 12, 2024 at 9:33 AM Robert Haas <robertmhaas@gmail.com> wrote:
> On Mon, Mar 11, 2024 at 11:11 PM Jingxian Li <aqktjcm@qq.com> wrote:
> > Your version changes less code than mine by pushing the nowait flag down
> > into ProcSleep(). This looks fine in general, except for a little advice,
> > which I don't think there is necessary to add 'waiting' suffix to the
> > process name in function WaitOnLock with dontwait being true, as follows:
>
> That could be done, but in my opinion it's not necessary. The waiting
> suffix will appear only very briefly, and probably only in relatively
> rare cases. It doesn't seem worth adding code to avoid it.

Seeing no further discussion, I have committed my version of this
patch, with your test case.

Thanks for pursuing this improvement!

-- 
Robert Haas
EDB: http://www.enterprisedb.com



Commits

  1. Fix typo and comments related to the recent no-wait lock improvements

  2. Allow a no-wait lock acquisition to succeed in more cases.